Skip to main content

Getting Started

This guide will walk you through the process of setting up your account, creating a notebook, loading and extracting data, generating embeddings, creating a view for vector search, creating a model, and querying the model. Let's get started!

Sign-Up​

Fill in your "Email" and "Organization" to complete the sign-up process.

Sign Up GIF

Sign Up GIF

Once your account is created, you'll be taken to the LangDB dashboard, where you can access your notebooks and start working with your data.

Create a Project​

Create Project

Create a new project which will contain your applications. You can also connect to your database or use LangDB's Clickhouse database.

Create a new Notebook​

To create a new notebook, click on the "New" button on the dashboard.

Select a notebook

If you want to start with a pre-configured notebook template, you can choose to clone an existing notebook by selecting the desired notebook from the list and clicking on the "Clone" button .

Clone notebook

Loading and Extracting Data​

With your notebook clonsed, you can now start running the cells to load and extract data. LangDB supports various data sources, including CSV files, PDFs, and more.

Loading and Extracting Data

Generating Embeddings​

LangDB allows you to generate embeddings from your data, which are numerical representations of the text that capture semantic meaning. To generate embeddings, create a "Provider" and a "Embeddding Model", we are using OpenAi here. Generating Embeddings

To enable efficient vector search on your embeddings, you need to create a view. Here, we are using "Cosine Distance" to rank the results.

Create View

Once the view is created, you can perform vector searches on your embeddings to find similar data points.

Creating Prompt and Model​

LangDB allows you to create prompts and models for generating text based on your data.

Model Creation

Querying Model​

Once the "Model" is created, we can query the model using SQL.

Model Query